“…Third, a study examining the GHGD–COC nexus in the UK context is virtually nonexistent. Fourth, from a methodological perspective, all prior GHGD–COC studies have used the conventional linear regression methods, including the ordinary least squares (OLS) model (Albarrak et al, 2019; Kumar & Firoz, 2018; Li et al, 2017; Maaloul, 2018), the fixed‐effects regression model (Raimo et al, 2020; Trinks et al, 2017), the two‐stage least squares (2SLS) regression model (Bui et al, 2020) and the three‐stage least squares (3SLS) regressions (He et al, 2013; Lemma et al, 2019) (see Table 1). These traditional linear regression models summarise the average association between different dependent and independent variables on the basis of the conditional mean function E ( y | x ), which provides a partial understanding of the studied relations (Cobb‐Clark, Kassenboehmer, & Sinning, 2016).…”